Abstract 1254: Mechanistic interrogation of pre-treatment low dose aspirin effects in HER 2 positive breast cancer

Background: Prior data (Barron et al. Cancer Res. 2014 74:4065-77) suggests that pre-diagnostic exposure to aspirin can have significant effects on breast tumor biology and patient outcome. It has been proposed that aspirin inhibition of COX-2 may suppress lymphangiogenesis and metastasis (Karnezis et al Cancer Cell. 2014. 21:181-95). Here, we sought to recapitulate pre-diagnostic aspirin exposure in rodent models of Her2+ breast cancer and elucidate mechanisms of action. We also determined the effect of aspirin on tumor stroma, using a co-culture system of human tumor and mesenchymal stem cells (MSC). Methods: NOD/SCID mice were orthotopically implanted with Her2+ MDA-MB-231 or HCC1954 cells. 48hr later, animals began a daily low dose [30mg/kg or 120mg/kg] of aspirin, until tumors reached 250mm3. They were then resected. 3 weeks later, HCC1954 implanted animals were treated with trastuzumab (15mg/kg) and paclitaxel (5mg/kg) for 6 weeks. Primary tissues were analysed by immunohistochemistry to assess VEGF-C, -D, COX-2, LYVE1 and CD31. RNAseq was performed on tumours to identify aspirin perturbed molecular pathways. To determine the stromal response to aspirin, patient derived MSCs were cultured either alone or with HCC1954 cells and exposed to aspirin (2.5 or 7.5mM). Secreted VEGF-C was quantified. A tubule formation assay was performed to determine the impact of aspirin on angiogenesis. Pro-angiogenic protein expression was investigated using a human angiogenesis array platform. Results: A significant delay in tumor growth was observed in both tumor models following aspirin treatment (p